Introduction
In this article I’d review Gwent: The Witcher Card Game Community Patch August 2026 generated by democratic voting via Balance Council tool. I’d sketch the situation before the patch, analyze the changes one-by-one and then provide summary and a general review.
Check out the discussion of balance changes on Reddit.
It was the 27th council we formed Coalition with Shinmiri and published our balance suggestions on Reddit. I also welcome you to check out ‘Balance Council‘ page on leriohub if you haven’t yet – there is a lot of practical information and my Council philosophy. There is also an analysis of how Balance Council works in practice: What Gwent Balance Council Does? Successes, Failures and Democracy Analogies
Before Patch
In August Season 2026 I played over 400 games on Gwent Pro Ladder, which gives me a good view on ladder meta at 2.5k+ fMMR.
August Season 2026 experience was marked by overpowered Mobilization decks, out of which after first weak of the season Golden Nekker Arbalests proven especially deadly. GN Arbalests were already 2.6k+ deck in the preceeding season (July) in the hands of player -Innuendo-, who even published a guide on playgwent decks. Mobilization provision buff popularized this deck in a barely changed form (main netdeck). Mobilization leader used on Arbalest is 10 points of tempo and ~20 points in a long round assuming opponent can’t deal with 2 threats in one turn. Also 20 points is just a starter which doesn’t include extra synergies (removal value, Onager, Revenants…) or protection value. Its a “bleed out or lose” leader when matched with most normal decks, which is unhealthy by definition when lost R1 = 100% lost game. Strong Mobi GN forced use of a handful of NG/SY meta decks which excel at control – a great deal of archetypes was cut out unless willing to accept ~10% win chances when matched against Mobi GN. In the highest moment of the season due to GN Mobi NR had 3%+ overall winrate advantage over 2nd faction. I’ve seen many new nicknames on high ladder running GN Mobi, players who so far struggled to climb high with other decks.
I hard criticized Mobilization buff in the patch review and my predictions proven full correct. It is important to understand the roots of the buff. For sure 2.6k+ fMMR GN Arbalests going off the radar were one factor, but not the most important. Also the concept of buffing leaders to avoid real nerfs or give more deckbuilding freedom to all of them isn’t the crucial part. The main one is exclusive inference made from ladder meta rather than basic math. On top of spotting less played stuff, a deeper analysis is needed: is the card/leader less played because weak, or just underrated? Is the card/leader becoming top tier good or bad for the meta? There always would be less played stuff, not always weaker and not always even spread of popularity between archetypes is possible or good for ladder experience. Less played doesn’t equal unplayable. Can’t assume if majority of ladder doesn’t play something then its weak; need to make objective evaluation before taking balance council decisions. Unless these is understood, same mistake would be repeated very soon.
Second deck which stood out in the ladder experience was Battle Trance Runestone Spam Otkell. It was obvious that Runestone buff is a funky one, possible to get reverted soon. In practice the variance of Runestone proven very high indeed with extra Freya’s Blessing for Otkell not being a niche event. Runestone Spam was limited in ladder performance by Mobilization Golden Nekker, against which the matchup is unfavored due to lack of control.
Third deck, which didn’t get buffed in BC patch, but nevertheless was characteristic for August Season 2026 is Sihil NG. I’ve met it played last months as well, but only now a real boom happened, which many players getting high scores. Imprisonment Sihil Assimilate was mainly popular in China, but this month also got adopted for example by Kerpeten to achieve 2.6k+ score. While the list is not bad overall, this season with prevalent Mobi GN was especially rewarding for Sihil NG.
Fourth new element of the season were Fruits decks with Big Bois and Imlerith.
2.5k+ pro ladder meta:
MO | Fruits Midrange, Vampires Devo/Non-Devo, All-unit Nova Fruits, Tatterwing, Midrange Arachas, Carapace Relicts, White Frost (more early in the season)
NG | Sihil Assimilate, Formation Soldiers (late season), Enslave 6 Damage, Enslave 6 Assimilate, Renfri, TD Constructs, Hyperthin Gascon, Aristocrats
NR | Mobilization GN Arbalests, …
SK | Runestone Spam Alchemy, Pirates, BoG Warriors, Svalblod Sigvald
ST | Midrange Gift, Precision Strike Dwarves Swarm, Dragons
SY | Pockets Crimes, Classic Bounty
10.835mmr from lavk945 is the new all-time total MMR highscore! The previous one was 10.810 from Pajabol in the January Season 2025 of Gwentfinity (profile pic). Before Gwentfinity the highest MMR was 10.809 by Kams (lookup alltime highscores chart for pre-Gwentfinity times). Congratulations!
To achieve 10.800+ score one needs 2.700+ average score on factions. For four consecutive seasons the winner had very high score, but lavk’s result absolutely stands out; also almost 300 MMR advantage over the 2nd player is unreal. Probably Lavk is also the first player in the history to achieve three 2.7k+ scores in one season; the results are neatly spread between factions, none stands out.
It looks that Gwentfinity reached the stage where very high winning scores are possible. The best environment for a highscore is when the skill gap between 1st and next players is substantial and the chasing players play a lot of games. The worst enviroment is when there are many strong players around who play low-to-average amount of games, constantly being underrated with respect to the leader while draining mmr from the game spammers. ELO is relative – it reflects the odds to win a game against average ladder opponent queued in the given season rather than absolute quality of play. So on one hand the outstanding score of Lavk is inspiring and says that skill matters, on the other it indicates that less high-profile players are part-time involved on Pro Ladder.
No offense I hope, but privately I use the position of player KronoSex as an index. The lower (higher place) it is, the less competitive the season. This year so far it was 3 | 7 | 23 | 7 | 11 | 5 | 2 | 3 .
Balance Council Results
(++) great change
(+) good
(/) acceptable
(-) bad
(–) very bad
Nerf Brackets
-1 Power Decreased
Harald an Craite (6 -> 5)
(/) Placeholder. Evolving cards are one of acceptable placeholders in my book, although they make the game very slightly worse for very niche situations where we want to play 1st form in Round 1.
Usurper: Officer (6 -> 5)
(/) Placeholder.
Braathens (4 -> 3)
(-) This change is plausible in the long term context (as the power level of meta cards is meant to drop and there are few power nerf targets), but in my opinion the short term outlook is pretty bad – with E6 Assimilate netdeck being less frequent last season at high mmrs, Braathens 3-power would be viewed as a premature overnerf and likely instant reverted by casuals due to how popular E6 Assim is. Also a nerf to Braathens discentivizes Assimilate builds in favor of rather uninspired midrange NGs. I’d much prefer a power nerf to one of NG Soldiers, which however proved top cut only late in the season (2724 fMMR from Lavk).
I think Braathens may be another classic case of a NG card which has as many lovers as haters and therefore bounces back and forth in the Balance Council, just like Coup de Grace does lately.
Eldain (5 -> 4)
(++) Explanation
Sigi Reuven (4 -> 3)
(++) Explanation
Eveline Gallo (6 -> 5)
(+) Balancewise surely deserved. Eveline would still be playable, but to be used with some protection like Blacksmith, Stratagem or Candle.
One can argue that provision nerf was better for higher survivability, but then there would be barely any card to nerf by power but placeholders.
Bride of the Sea (4 -> 3)
(-) The concept of this nerf was to touch very gently the Stribog Runestone replayer rather than Runestone itself. As one could have expected, combined Ofir recommendation and casual sentiment in the community lead to Stribog nerf, which made Bride power nerf look out of place.
I think the concept was fundamentally wrong; 4-cost Stribog forces all Alchemy decks into RNG direction and such a state just couldn’t hold for long, even if somehow an instant revert got avoided.
Trollololo (5 -> 4)
(-) Out of 6 Mobilization GN nerfs which happened in this patch, this one in my opinion is least essential to freeze down engine overload GN type of decks. Trollololo is almost a winmore card in Arbalest GN and doesn’t work with many other types of bronze spam Mobi. I think this nerf would likely lead to Trollololo not being played again as a worse kind of Onager.
Musicians of Blaviken (2 -> 1)
(+) Explanation (personally I preferred to suggest this change later because Sihil is very popular right now and Musicians give a guaranteed deathblow from blue coin with no chance of counterplay)
Eternal Eclipse Deacon (4 -> 3)
(+) Last time Eternal Eclipse Deacon was at 4-power the ladder experience was pretty bad for me. This month I’ve met few Cultists players, who sometimes tried own version of the deck and weren’t that successful. I think that the nerf is reasonable but I wouldn’t mind if it didn’t happen yet either.
+1 Provision Increased
Lippy Gudmund (12 -> 13)
(–) Unjustified balancewise; Lippy and Shupe have already been nerfed. I’ve met Lippy very rarely on the ladder last month. Lippy ability is unique and important for the variety of deckbuilding and gameplay concepts in Skellige.
Shani (9 -> 10)
(++) Explanation
Vernon Roche (9 -> 10)
(++) As every card which can be buffed by power decrease (Rainfarn, Ivar…), Vernon Roche has a rich BC story. Started as 3/9, then in BC => 2/9 => 1/9 => 1/10 => 1/9 => 1/10 => 1/9 => 2/9. Moreover to keep Roche in sub-10 provision region, a card which can replay him: Hanmarvyn’s Blue Dream got nerfed to 11-cost.
Vernon plays 2 cards from the topdeck, which in raw value should be no less than 2×7 (weakest bronze value) – Roche power. So 2/9 Roche in raw points was 12+ for 9. Okay, but not overwhelming. To max Roche value we need to benefit from playing two cards in one turn or setup these two cards to work well together. Example of the first is engine overload – if extra engine lives through the full round, then Roche is more like 22+ for 9. Example of the second is Priestess + Onager/Tridam Infantry combo, but you can also think of other, more savage plays like Regis + Ornate Censer.
In the context of Mobi GN prevalence, Roche nerf is welcomed. I think there is a perfect justification to keep this card out of GN range due to doubling down on engine overload effect. Not only explicitly Arbalest decks could be a problem. Moreover Roche at 10-cost unlocks Hanmarvyn Blue Dream for provision buffs.
Hefty Helge (9 -> 10)
(-) While even after the last 8=>9 cost nerf to Helge the Damage Enslave 6 netdeck with Joachim and Emhyr was still quite popular, this one is surely a step too far. Yes, Helge can win games when not countered, but otherwise it’s a vulnerable card, which also can be efficiently bled out as Helge plays for 4 tempo.
Ethereal (8 -> 9)
(++) Explanation
Avallac’h: Sage (7 -> 8)
(++) Explanation
Tunnel Drill (7 -> 8)
(++) Plausible and widespread nerf – Drill is included in many SY decks, from Crimes to Bounty. Especially when protected with Conjurer’s Candle, Drill can singlehandedly control the board with a little support of Crimes, for example with Hysteria Drill can remove a unit with power up to 9 if already sticked on the board. Hurts, but have to hurt or no nerfs ever will stick.
Kingslayer (5 -> 6)
(++) At high mmrs Mill was very rare and Kingslayer as a random carryover card also barely met. I think that Kingslayer nerf wasn’t that essential to happen this month, but nevertheless is fine. If Mill lovers would like to buff Kingslayer next patch I much recommend +1 power direction.
Lyrian Arbalest (5 -> 6)
(/) The main concern for the next season is Mobilization GN not being any close to current power after swapping one or two cards. To achieve this, Lyrian Arbalest nerf is acceptable, even though the card itself outside the context of opening play with Mobilization leader becomes weak. I think that Arbalest even if unplayed after nerf shouldn’t be reverted immediately, because at this point the coupling with Mobilization is real; god forbid going for power buff.
Stribog Runestone (4 -> 5)
(++) From the ladder experience it was clear that the Runestone package couldn’t stay in the current form and needs to be nerfed directly back or via replayers, which are Crow Clan Druids and Bride of the Sea. The direct revert lowers variance; Stribog at 5-cost is not terrible, but slightly sub par. The other route would have kept high variance in the game and Alchemy strongly coupled with the Create mechanic, which shouldn’t be the case.
Buff Brackets
+1 Power Increased
She Who Knows (11 -> 12)
(-) I much preferred the provision buff direction. She Who Knows is a red coin polarized card – could be played for reach with no possible answer other than damaging units to the point Sabbath condition is impossible to be satisfied. Outside of possible effective +3 power buff (if SWK sticks in each round), power buff improves the mentioned reach too. The ceiling of unanswered SWK is outstanding, which further promotes archetypes with access to multiple control tools.
The Great Oak (9 -> 10)
(/) While rarely played I believe Oak was already a decent card. Oak supports pointslam + control decks which can swarm, like many Precision Strike builds.
Geralt: Igni (2 -> 3)
(/) Igni was rare, but a considerable card. Power buff would incentivize playing it more for surprise value (which would probably be the main effect), but also Igni becomes a better payoff in dedicated builds like late control Guerilla.
Gimpy Gerwin (5 -> 6)
(+) Good change. I’d prefer to see something more impactful, but buff was necessary to keep Gimpy safe from nerf.
Prince Stennis (5 -> 6)
(/) While I’m happy that the starter set Prince Stennis isn’t a placeholder, the whole process of bringing Stennis to playable state is slow and in my opinion shouldn’t have been taken. There is Nenneke in the starter set with more synergies, closer to playable state and for example capable of setting up Geralt:Igni buffed in this patch.
Greater Brothers (11 -> 12)
(/) Okay buff, but not really necessary. The amount of insanity damage for cards like Arena Endrega wouldn’t change after this power buff, only Greater Brothers get more tempo and reach.
Vanhemar (3 -> 4)
(+) Not the most awe-inspiring change, but justified in my opinion. Vanhemar is usually coupled with Imposter, which makes this leader less flexible – you have to decide between locking a threat vs locking the best Vanhemar target. For this reason in practice Vanhemar would require support of other lock cards. The less flexible leader is the hidden cost of Vanhemar if you’d like to compare it with Leo Bonhart or other tall removals. For example if you use Imposter on Mourntart to maximize Vanhemar value, then your leader may be worth 2 raw points!
Elven Swordmaster (4 -> 5)
(-) One of the bad changes which were hanging over the Balance Council like Damocles sword. It just was doomed to happen. 5 power Swordmaster is a very good proactive bronze which supports efficient damage cards like Backup Plan. Swordmaster could be played pretty much anywhere.
I think that Swordmaster may not feel as strong as it really is because of Scoia’tael mirrors where both Trickster and buffed Trained Hawk can trade up big time against Swordmaster.
Trained Hawk (3 -> 4)
(++) After power buff Trained Hawk becomes a decent card, very efficient in a long round against row locked units. As Beast tag is rather common in Harmony, otherwise Trained Hawk would feel just okay. Hawk can also be used for self-movement with units like Weeping Willow or Forest Whisperer, especially when Call of Harmony leader is used over Guerilla Tactics.
Heymaey Herbalist (1 -> 2)
(/) Wholesome unplayed card, but I think that but for the art and voicelines it wouldn’t bring much new. Could be combined with Bear Witcher/Spearmaiden/Havrue Singer.
-1 Provsion Decreased
Mobilization (16 -> 15)
(++) Check out my comment in the Before Patch section and previous review.
One Night at the Passiflora (13 -> 12)
(/) Passiflora wasn’t much played in Syndicate. The provision buff is acceptable to give it more attention, but in my opinion not necessary balancewise. Sly Seductress is 5 power now. Passiflora is mainly used to force Round 1 in decks like 2x Scenario Hidden Cache.
Vilgefortz: Renegade (12 -> 11)
(++) Explanation
Roche: Merciless (11 -> 10)
(+) If Deathblow condition is satisfied, Roche: Merciless in Round 1 summoning 2x Blue Stripes from the deck would be 6+3*5 = 21 points for 10+2*2 = 14 cost, plus 5 points contribution to carryover with Pavetta. That’s ceiling value though and oftentimes getting a deathblow soon enough would be troublesome. Also giving Zeal to Commando after deathblow is a hidden cost.
I don’t love the current Blue Stripes archetype, especially in the context of Renew Pavetta, where the game often comes down to round 1 draws. Yet it can be interesting to see Roche playable as a midrange card with 2x thinning in Zeal Damage decks, while Stripes carryover abuse is balanced separately.
King Bran (11 -> 10)
(+) I actually like this change. Bran is associated with Raids Warriors and therefore many players would view it as a “powercreep” buff to an already figured out archetype. Without Warlords though, Bran is hardly justified right now – for example Bran is omitted in Eist Warriors netdeck because of being far too expensive.
I think Bran should get a little bit better in decks not running Warlords. I doubt old netdeck would be much of a problem after this small buff; otherwise it is easy to address PF Raids directly if necessary.
Kolgrim (10 -> 9)
(-) Kolgrim is not outstanding right now with many decks thinning very well, therefore buff is plausible balancewise. I much more liked +1 power direction though, which keeps Kolgrim out of Renew and Golden Nekker range (I got 2666 fMMR with GN Clog), while Kolgrim doesn’t die to An Craite Longship or Naval Supremacy. Also balance is not everything; cards like Kolgrim which are abused in a figured out, linear way just to see the opponent suffer due to missed answer shouldn’t be incentivized.
Kayran (7 -> 6)
(++) Finish of 2-step change to bring Kayran out of Penitent range.
Bloody Flail (6 -> 5)
(++) I played a bit with this niche Warfare card and it was better than its common opinion. With Kerack Frigate or Poor Fucking Infantry buffed in the Balance Council its easier to provide Soldiers condition than before. Blood Flail can be used for example in Stockpile decks with Kerack Frigate, where it would also trigger Ressuply on Siege Engines.
Saov Ainmhi’dh (6 -> 5)
(–) Saov would become included as a low cost carryover possibility (especially from red coin) in each deck which plays enough specials to get it out in two rounds. Saov can also now be created from Filavandrel (not guaranteed without Runemage), which however is not very provision efficient unless manage to get a strategic advantage from carryover in each round.
Phooca (5 -> 4)
(++) Good to see this 2-step change finished.
Summary
All changes ordered by the number of votes and attributed to factions and voting coalitions could be found in the BALANCE COUNCIL RESULTS SHEET.
Out of 12 changes we recommended with Shinmiri this month, 8 got through – thanks for your support! The missing changes are all buffs which we recommended as the only coalition: Kolgrim +1p, Cintrian Artificer +1p, Piercing Missile -1c and Regis -1c. On the other hand original nerfs: Sigi Reuven, Eldain and Ethereal made the cut on the very last slots and Musicians as the 6th. It seems that our recommendations had less influence this time. Not sure what was the cause given that this month changes were generally more favorably received on Reddit and Youtube than usual. Feedback welcomed.
Golden Nekker Moblization got addressed by 6 nerfs, which is a lot, but better too much than too few. One would say that Mobilization and Lyrian Arbalest nerf were enough (what if not though?) and the rest unnecessarily impacts too many decks, the others would say that all changes but Arbalest and Trollololo are welcomed as both of these cards become weak. I think that there is no reason to treat Shani or Vernon as overnerfs as in the span of a few months those are amongst considerable downgrades anyway. I’m relatively happy about this part of the balance council and hope for no immediate reverts.
As expected, Scoia’tael got the most impactful buffs by a lot. Trained Hawk, Great Oak, Saov and Elven Swordmaster all would be played. Saov is supposed to be included in any ST deck running a couple of specials. Elven Swordmaster also would be run almost everywhere, from Dragons and Elves to Midrange Gift.
September Season 2026 meta is supposed to include above all Nilfgaard Soldiers (unnerfed 2.7k+ deck) and Midrange ST decks. Soldiers not getting addressed is in my opinion one of the main failures for the incoming month, but also they got prevalent only in the late second half of the season.
A recurring part of balance council discussion is blaming disliked changes on “dictatorship” compared with an ideal democratic process with no established coalitions. If we look deeper though, both MetallicDanny and Violet.Kiramman recommendations have strong democratic foundations. MD admittedly often picks what he personally is not a fan of, but got highly upvoted in his community. For Violet.Kiramman you could see the results of balance council survey exactly converging with recommendations. Yes, placeholders are highest upvoted of all cards there. Without alignment the placeholders wouldn’t be possible, so one can say its the fault of coalitions,… yet that’s what most people want. Simultanously in the other parts of the internet people hate placeholders more than overnerfs. Remember: in the Balance Council the number of supporters matters, not the number of people against a change. Unpopular changes can’t even be forced through by the authorities; worst case they are popular only locally.
What New To Try?
MO | Threat Overload / Relicts with She-Who-Knows, Succubi with Kayran, Vampires with Geralt: Igni
NG | Sihil Assimilate w/Vilgefortz: Renegade, Midrange Imposter w/ Vanhemar, Recruits w/Gimpy Gerwin
NR | Stockpile Swarm w/Bloody Flail, Blue Stripes w/Roche: Merciless, Zeal Draug w/Roche: Merciless
SK | Non-Raids Veterans decks with King Bran, Sihil Raids w/King Bran
ST | Midrange Slop w/Saov, Swordmaster, sometimes Oak, Harmony w/Trained Hawk, Guerilla Control w/Geralt: Igni & Elven Swordmaster
SY | 2x Scenario Cache, Bonded SY w/Passiflora, Greater Brothers power shouldn’t matter much for OtB Insanity w Arena Endrega, Blindeye Apothecary and Iris: Shade.
Closure
September 2026 Patch is mediocre; if you had expectations for each change to make sense and click well with the others then even bad. I predict a boring season with prevalent NG Soldiers and Midrange ST, so I can only be positively surprised. Mya-mon is back to Gwent so maybe would cook something from a different direction.
See you on ladder!
lerio2